The swift growth of e-commerce has revolutionized the fashion sector, making it easier to shop online while also introducing challenges such as a lack of personalization, obstacles in product discovery, and the inability to see how clothing looks before making a purchase. This initiative tackles these challenges by creating a Fashion Recommendation System and Virtual Trial Room that integrate artificial intelligence (AI) and computer vision to improve user experience and engagement in online shopping. The paper presents Fashion Recommendation System which utilizes deep learning and machine learning strategies to offer users precise and tailored clothing suggestions. The system is enhanced by the Virtual Trial Room, which allows customers to preview clothing items before purchase. Utilizing computer vision technologies and JavaScript, the trial room enables users to place specific clothing pieces onto a digital mannequin. Distinct mannequins for both male and female users ensure inclusivity and precise visualization. This functionality significantly reduces uncertainty regarding fit and style, thereby decreasing the number of returns for online purchases. The architecture of the project features an intuitive frontend developed with HTML, CSS, and Bootstrap, while the backend, created with Flask, ensures seamless integration of recommendation and visualization features. For data management and user authentication, PostgreSQL is used as the database to store recommendations securely. Performance testing revealed that the system efficiently processes images and generates recommendations in under three seconds, demonstrating its scalability and real-time responsiveness.